Method and device for recommending cargo sources to drivers

Abstract

The present application discloses a method and device, an electronic device, and a computer-readable medium for recommending cargo sources to a driver. The method comprises: determining a plurality of tags and their weights, wherein the plurality of tags at least comprise a time matching tag and a location matching tag, wherein the time matching tag is used to indicate the matching degree between the driver and the cargo source in the dimension of time, and the location matching tag is used to indicate the matching degree between the driver and the cargo source in the dimension of location; determining a score for each of the plurality of tags according to basic information of the driver, transportation plan information, transportation tool information, cargo source information, and transportation demand; calculating a matching score between each cargo source and the driver according to the score of each tag and its weight; and recommending cargo sources to the driver according to the matching score, thereby solving the problem in the prior art that the transportation tasks of the carrier group are not continuous enough and the empty driving rate is high, effectively reducing the empty driving rate and operating costs, and arranging the driver's future work plan in advance.

Description

Technical Field

[0001] The present application generally relates to a recommendation algorithm, and more particularly to a method and device, an electronic device, and a computer-readable medium for recommending cargo sources to a driver. Background Art

[0002] With the implementation of my country's network infrastructure, the Internet has become an inseparable part of people's lives. Various Internet-based applications have sprung up like mushrooms after rain. Network technology has deeply transformed all aspects of traditional industries. Similarly, Internet technology has also opened up new horizons for the logistics industry. How to give full play to the advantages of Internet technology, integrate idle social transportation resources, and improve logistics efficiency has become an urgent need for the logistics industry.

[0003] In the logistics industry, especially for trunk trucks and coastal bulk cargo, there are a lot of planned logistics scenarios. For example, it is very common for manufacturers to release annual or quarterly logistics plans and transport goods as planned within a certain period. On the other hand, most carriers undertake business on a one-way basis, which results in insufficient continuity of transportation tasks, high empty driving rate and high operating costs. Summary of the invention

[0004] In view of the defects of the prior art, the present application provides a method and device, an electronic device, and a computer-readable medium for recommending sources of goods to drivers, which improve many problems existing in the prior art.

[0005] The present application provides a method for recommending sources of goods to drivers, comprising: obtaining source information and transportation requirements of multiple sources of goods; obtaining driver information and transportation tool information, wherein the driver information includes basic driver information and transportation plan information; determining multiple tags and their weights, wherein the multiple tags include at least a time matching tag and a location matching tag, wherein the time matching tag is used to indicate the matching degree between the driver and the source of goods in the dimension of time, and the location matching tag is used to indicate the matching degree between the driver and the source of goods in the dimension of location; determining the score of each of the multiple tags according to the basic driver information, the transportation plan information, the transportation tool information, the source of goods information and the transportation requirements; calculating the matching score between each source of goods and the driver according to the score of each tag and its weight; and recommending sources of goods to drivers according to the matching scores.

[0006] Furthermore, the multiple labels also include at least one of the following: a transport model matching label, a transport load matching label.

[0007] Furthermore, after obtaining the driver's transportation plan information, the driver's free time information is determined based on the transportation plan information; after obtaining the cargo source information and transportation demand, the loading time and place of the cargo source's transportation demand, and the unloading time and place of the transportation demand are determined; the scores of the time matching tag and the place matching tag are determined based on the driver's free time information, the loading time and place of the cargo source's transportation demand, and the unloading time and place of the transportation demand.

[0008] Furthermore, the time matching tag includes two types of time matching tags, wherein the first type of time matching tag is used to indicate whether the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, and the second type of time matching tag is used to indicate when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, the length of the time interval between the two endpoints of the time interval of the transportation demand of the cargo source and the two endpoints of the free time interval of the driver.

[0009] Furthermore, when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, the score of the location matching tag is calculated, wherein the location matching tag is used to indicate the distance between the loading location of the transportation demand of the cargo source and the driver's location at the left end point of the driver's free time interval, and the distance between the unloading location of the transportation demand of the cargo source and the driver's location at the right end point of the driver's free time interval when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver.

[0010] Furthermore, the method also includes optimizing the weight of each label. Optimizing the weight of each label specifically includes: after recommending the source of goods to the driver according to the degree of matching, recording the driver's evaluation information on the source of goods; optimizing the weight of each label according to the driver's evaluation information on the source of goods.

[0011] Furthermore, the method of optimizing the weights of various labels based on the driver's evaluation information on the cargo source specifically includes: adjusting the weights of the labels, recalculating the matching score between each cargo source and the driver based on the score of each label and its adjusted weight, and re-recommending the cargo source based on the matching score; obtaining the driver's score for the re-recommended cargo source; calculating the recommended deviation parameter based on the deviation between the matching score between the cargo source and the driver and the driver's score for the re-recommended cargo source; and determining the weight of the label when the recommended deviation parameter is the smallest, as the weight of the optimized label.

[0012] The present application also provides a device for recommending sources of goods to drivers, characterized in that it includes: a first acquisition unit, used to obtain source information and transportation requirements of multiple sources of goods; a second acquisition unit, used to obtain driver information and transportation tool information, the driver information includes driver basic information and transportation plan information; a first determination unit, used to determine multiple tags and their weights, the multiple tags at least include a time matching tag and a location matching tag, the time matching tag is used to indicate the matching degree between the driver and the source of goods in the time dimension, and the location matching tag is used to indicate the matching degree between the driver and the source of goods in the location dimension; a second determination unit, used to determine the score of each of the multiple tags according to the driver basic information, the transportation plan information, the transportation tool information, the source of goods information and the transportation demand; a calculation unit, used to calculate the matching score between each source of goods and the driver according to the score of each tag and its weight; a recommendation unit, used to recommend sources of goods to drivers according to the matching score.

[0013] Furthermore, the multiple labels also include at least one of the following: a transport model matching label, a transport load matching label.

[0014] Furthermore, after obtaining the driver's transportation plan information, the driver's free time information is determined based on the transportation plan information; after obtaining the cargo source information and transportation demand, the second determination unit determines the loading time and place of the cargo source's transportation demand, and the unloading time and place of the transportation demand; and the scores of the time matching tag and the place matching tag are determined based on the driver's free time information, the loading time and place of the cargo source's transportation demand, and the unloading time and place of the transportation demand.

[0015] Furthermore, the time matching tag includes two types of time matching tags, wherein the first type of time matching tag is used to indicate whether the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, and the second type of time matching tag is used to indicate when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, the length of the time interval between the two endpoints of the time interval of the transportation demand of the cargo source and the two endpoints of the free time interval of the driver.

[0016] Furthermore, when the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver, the second determination unit calculates the score of the location matching tag, wherein the location matching tag is used to indicate the distance between the loading location of the transportation demand of the source of cargo and the driver's location at the left end point of the driver's free time interval, and the distance between the unloading location of the transportation demand of the source of cargo and the driver's location at the right end point of the driver's free time interval when the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver.

[0017] Furthermore, the device also includes an optimization unit for optimizing the weights of each label, and the optimization unit specifically includes: a recording subunit for recording the driver's evaluation information on the source of goods after recommending the source of goods to the driver according to the degree of matching; an optimization subunit for optimizing the weights of each label according to the driver's evaluation information on the source of goods.

[0018] Furthermore, the optimization subunit specifically includes: an adjustment module, used to adjust the weight of the label, recalculate the matching score of each cargo source and the driver according to the score of each label and its adjusted weight, and re-recommend the cargo source according to the matching score; an acquisition module, used to obtain the driver's score for the re-recommended cargo source; a calculation module, used to calculate the recommended deviation parameter according to the deviation between the matching score between the cargo source and the driver and the driver's score for the re-recommended cargo source; a determination module, used to determine the weight of the label when the recommended deviation parameter is the smallest, as the weight of the optimized label.

[0019] The present application also provides an electronic device, characterized in that it includes: one or more processors; a storage device for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the above method.

[0020] The present application also provides a computer-readable medium having a computer program stored thereon, wherein the program implements the above method when executed by a processor.

[0021] The technical solution of the present invention uses time matching tags to match drivers and cargo sources in the dimension of time, and uses location matching tags to match drivers and cargo sources in the dimension of location. When calculating the matching score of each cargo source and driver, the time matching and location matching factors are added, so as to screen out transportation tasks that are well connected with the driver's existing transportation plan in the dimensions of time and location. This solves the problem in the prior art that most carriers undertake business on a one-way basis, the transportation tasks are not continuous enough, and the empty driving rate is high, effectively reduces the empty driving rate and operating costs, and arranges the driver's future work plan in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly introduced below. Obviously, the drawings in the following description only relate to some embodiments of the present disclosure, but are not intended to limit the present disclosure.

[0023] Figure 1 A flow chart of a method for recommending a source of goods to a driver provided in an embodiment of the present application;

[0024] Figure 2-1 , 2-2 , 2-3, and 2-4 are schematic diagrams of matching a driver with a cargo source provided in an embodiment of the present application;

[0025] Figure 3 A flow chart of a method for recommending a source of goods to a driver provided in an embodiment of the present application;

[0026] Figure 4 A flow chart of a method for recommending a source of goods to a driver provided in an embodiment of the present application;

[0027] Figure 5 A schematic diagram of a computing framework of a system for executing a method for recommending cargo sources to a driver provided in an embodiment of the present application;

[0028] Figure 6 A schematic diagram of a device for recommending cargo sources to drivers provided in an embodiment of the present application;

[0029] Figure 7 A schematic diagram of a framework of an electronic device provided in an embodiment of the present application;

[0030] Figure 8 A block diagram of a computer-readable medium provided for an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution of the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings of the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0032] In the present application, unless otherwise specified or different meanings can be derived from the context, each term has the meaning generally understood in the art.

[0033] Transport plan, transport arrangement, transport task and task have the same meaning.

[0034] This application provides a method for recommending cargo sources to drivers, such as Figure 1 As shown, the following steps are included:

[0035] Step S101: Obtaining the supply information and transportation requirements of multiple supply sources.

[0036] The cargo source information and transportation requirements may include cargo information, cargo owner information, transportation start time, transportation end time, loading location (GPS longitude and latitude coordinates), unloading location (GPS longitude and latitude coordinates), transportation freight, etc. For example, cargo source 1 needs to depart from Jinan on January 2 and arrive in Tianjin on January 5. Cargo source 2 needs to depart from Taiyuan on January 7 and arrive in Xi'an on January 14. Cargo source 3 needs to depart from Jinan on January 1 and arrive in Baotou on January 4. Cargo source 4 needs to depart from Chengdu on January 8 and arrive in Hangzhou on January 12. Cargo source 5 needs to depart from Jinzhong on January 9 and arrive in Xianyang on January 11.

[0037] Step S102: Acquire driver information and transportation tool information, where the driver information includes basic driver information and transportation plan information.

[0038] The driver's basic information may include the driver's ID number, the driver's name, the driver's license information, the driver's qualifications, the driver's ID information, the driver's place of residence, the routes the driver frequently travels, the driver's historical score, etc.

[0039] The driver's transportation plan information refers to the driver's transportation task information, including transportation task number, transportation tool information, driver information, transportation start time, transportation end time, loading location (GPS longitude and latitude coordinates), unloading location (GPS longitude and latitude coordinates), task classification (history / ongoing / planned), etc. For example, driver Zhang San already has 2 tasks, task 1 is to depart from Tianjin on January 4, transport a batch of goods, and arrive in Taiyuan on January 8; task 2 is to depart from Xi'an on January 12, transport a batch of goods, and arrive in Guangzhou on January 17.

[0040] Means of transport refers to vehicles, ships and other means of transport that have the function of transporting goods. Basic information of means of transport includes the vehicle brand (such as license plate number, ship number), type of means of transport (such as vehicle model, ship type), load capacity, length, width and height, purpose of transport, loading and unloading methods, etc.

[0041] Step S103: Determine multiple tags and their weights, where the multiple tags include at least a time matching tag and a location matching tag. The time matching tag is used to indicate the matching degree between the driver and the cargo source in the dimension of time, and the location matching tag is used to indicate the matching degree between the driver and the cargo source in the dimension of location.

[0042] The inventor of this application found that the carrier group in the logistics industry has a high idle rate and high operating costs. The inventor thought that if the carrier's transportation tasks can be arranged in advance and the carrier's transportation tasks can be made continuous, it can not only solve the carrier's future work plan problem, but also effectively reduce the idle rate and operating costs. To this end, this application uses time matching tags and location matching tags to match drivers and sources of goods in time and space (location) respectively, so as to find sources of goods that are highly matched with drivers and recommend them to drivers.

[0043] like Figure 2-1 , 2-2 As shown in Figures 2-3 and 2-4, driver Zhang San is free today (assuming it is January 1st), currently located in Jinan, and has two scheduled transportation plans: Task 1 and Task 2. Task 1 is to depart from Tianjin on January 4th, transport a batch of goods, and arrive in Taiyuan on January 8th. Task 2 is to depart from Xi'an on January 12th, transport a batch of goods, and arrive in Guangzhou on January 17th.

[0044] Driver Zhang San's time from January 4 to January 8 is occupied by Task 1, and his time from January 12 to January 17 is occupied by Task 2. Driver Zhang San's free time is from January 1 to January 4, from January 8 to January 12, and after January 17. The most ideal cargo source recommendation is to recommend 3 cargo sources, which occupy the above 3 free time periods respectively, as shown in Figure 2-1. The departure point of cargo source 1 is the current location (Jinan), and the arrival point is the departure point of Task 1 (Tianjin); the departure point of cargo source 2 is the arrival point of Task 1 (Taiyuan), and the arrival point is the departure point of Task 2 (Xi'an); the departure point of cargo source 3 is the arrival point of Task 2 (Guangzhou), and the arrival point is a certain place (according to actual needs, it can be any place). Figure 2-1 The supply recommendations shown are the ideal scenario, with perfect timing and location.

[0045] like Figure 2-2 As shown in the figure, there is a cargo source 1, which needs to depart from Jinan on January 2 and arrive in Tianjin on January 5, cargo source 2 needs to depart from Taiyuan on January 7 and arrive in Xi'an on January 14, and cargo source 3 needs to depart from Guangzhou on January 15. In terms of time, these three cargo sources should not be recommended to driver Zhang San. Because if cargo source 1 is recommended to Zhang San, then Zhang San will depart from Jinan on January 2 to transport cargo source 1 and arrive in Tianjin on January 5, while Zhang San's original task 1 needs to depart from Tianjin on January 4, which delays task 1. Cargo source 2 should not be recommended to Zhang San either. Because cargo source 2 needs to depart from Taiyuan on January 7, while Zhang San is still on the way from Tianjin to Taiyuan on January 7, and can only arrive in Taiyuan on January 8. Cargo source 3 should not be recommended to Zhang San either. Because cargo source 3 needs to depart from Guangzhou on January 15, while Zhang San is still on the way from Xi'an to Guangzhou on January 15, and can only arrive in Guangzhou on January 17.

[0046] like Figure 2-3 As shown, there is a cargo source 1, which needs to depart from Jinan on January 1 and arrive in Baotou on January 4, cargo source 2 needs to depart from Chengdu on January 8 and arrive in Hangzhou on January 12, and cargo source 3 needs to depart from Dalian on January 17. These three cargo sources should not be recommended to driver Zhang San. Because if cargo source 1 is recommended to Zhang San, then Zhang San will depart from Jinan on January 1 to transport cargo source 1 and arrive in Baotou on January 4, while Zhang San's original task 1 needs to depart from Tianjin on January 4. Baotou is far away from Tianjin. After Zhang San arrives in Baotou on January 4, he cannot depart from Tianjin to perform task 1 on the same day, which delays task 1. Cargo source 2 should not be recommended to Zhang San either. Because cargo source 2 needs to depart from Chengdu on January 8, and Zhang San arrives in Taiyuan after completing task 1 on January 8. Taiyuan is far away from Chengdu, so he cannot rush to Chengdu to transport cargo source 2 on the same day. Cargo source 3 should not be recommended to Zhang San either. Because cargo source 2 needs to depart from Chengdu on January 8, and Zhang San arrives in Taiyuan after completing task 1 on January 8. Taiyuan is far away from Chengdu, so he cannot rush to Chengdu to transport cargo source 2 on the same day. Cargo source 3 should not be recommended to Zhang San either. Because cargo 3 needs to depart from Dalian on January 17, and Zhang San arrived in Guangzhou on January 17 after completing mission 3. Guangzhou is far away from Dalian, so he cannot rush to Dalian to transport cargo 3 on the same day.

[0047] like Figure 2-4 As shown in the figure, there is a cargo source 1, which needs to depart from Zibo on January 2 and arrive at Daxing on January 3, cargo source 2 needs to depart from Jinzhong on January 9 and arrive at Xianyang on January 11, and cargo source 3 needs to depart from Foshan on January 18. These three cargo sources are more compatible with the driver Zhang San in terms of time and location, and can be recommended to Zhang San. Specifically, Zhang San can depart from Zibo on January 2, load cargo source 1, arrive at Daxing on January 3, unload cargo source 1 and rush to Tianjin. Since the distance between Daxing and Tianjin is relatively close, he can arrive in Tianjin before January 4, depart from Tianjin on January 4, and arrive in Taiyuan on January 8 to perform task 1. Then rush to Jinzhong from Taiyuan, load cargo source 2 in Jinzhong on January 9, arrive in Xianyang on January 11, unload cargo source 2 and rush to Xi'an. Since the distance between Xianyang and Xi'an is relatively close, he can arrive in Xi'an before January 12. It departed from Xi'an on January 12 and arrived in Guangzhou on January 17 to carry out Task 2. It then rushed from Guangzhou to Foshan. Since the distance between Guangzhou and Foshan is relatively short, it can arrive in Foshan before January 18. It loaded cargo 3 in Foshan and departed on January 18.

[0048] Step S104: Determine the score of each tag in the multiple tags according to the basic information of the driver, the transportation plan information, the transportation tool information, the cargo source information and the transportation demand.

[0049] Step S105: Calculate the matching score between each cargo source and the driver according to the score and weight of each tag.

[0050] Step S106: recommending cargo sources to the driver according to the matching scores.

[0051] The weights of the time matching tags and the location matching tags may be set higher, so that the cargo sources that match the time and location can stand out, obtain higher matching scores, and thus be recommended to drivers.

[0052] Optionally, the multiple labels also include at least one of the following: a transport model matching label, a transport load matching label.

[0053] In this application, there is a type of label that has a dual screening function. Two screenings are implemented in the entire method of recommending sources of goods to drivers. The first screening process does not involve the score of the label, and the second screening process involves the score of the label. Only if you pass the first screening successfully, you will be eligible for the second screening. If the first screening fails, the matching process is terminated and it is directly determined that it is not recommended. This is like an exam, with 60 points as the passing score. If a student exceeds 60 points, he passes the first screening. In this case, the actual score of the student needs to be displayed (similar to the second screening); if a student scores less than 60 points, he does not pass the first screening. In this case, it is only necessary to show that the student has not passed, and the actual score is not displayed. For example, student Li Si's score is 88 points, student Wang Wu's score is 60 points, and student Zhao Liu's score is failing.

[0054] The load-carrying capacity of different means of transportation is different. For example, the means of transportation of driver 1 is vehicle 1, and the load-carrying capacity of vehicle 1 is 1.8 tons; the means of transportation of driver 2 is vehicle 2, and the load-carrying capacity of vehicle 2 is 6 tons; the means of transportation of driver 3 is vehicle 3, and the load-carrying capacity of vehicle 3 is 14 tons; the means of transportation of driver 4 is vehicle 4, and the load-carrying capacity of vehicle 4 is 100 tons; the means of transportation of driver 5 is vehicle 5, and the load-carrying capacity of vehicle 5 is 150 tons.

[0055] Different sources of cargo have different requirements for the load capacity of transportation vehicles.

[0056] For example, assuming that the weight of cargo source 1 is 110 tons, when matching according to the vehicle load matching degree, the weight of cargo source 1 exceeds the load capacity of vehicles 1, 2, 3, and 4, and fails to match with drivers 1, 2, 3, and 4. The matching degrees of the remaining tags are no longer calculated, and cargo source 1 will not be recommended to drivers 1, 2, 3, and 4. The weight of cargo source 1 is within the load capacity of vehicle 5, and it successfully matches with driver 5. The matching degrees of the remaining tags continue to be calculated, and cargo source 1 may be recommended to driver 5.

[0057] For example, assuming that the weight of cargo source 2 is 10 tons, when matching according to the vehicle load matching degree, the weight of cargo source 2 exceeds the load capacity of vehicles 1 and 2, and fails to match with drivers 1 and 2. The matching degrees of the remaining labels are no longer calculated, and cargo source 1 will not be recommended to drivers 1 and 2. The weight of cargo source 2 is within the load capacity of vehicles 3, 4, and 5, and successfully matches with drivers 3, 4, and 5. The matching degrees of the remaining labels continue to be calculated, and cargo source 2 may be recommended to drivers 3, 4, and 5. It should be noted that although cargo source 2 successfully matches with drivers 3, 4, and 5, the scores of these three matches in the dimension of vehicle load matching degree label are not the same. Among them, the score of cargo source 2 and driver 3 in the dimension of vehicle load matching degree label is the highest, and the score of cargo source 2 and driver 5 in the dimension of vehicle load matching degree label is the lowest.

[0058] Different sources of goods have different requirements for the models of transport vehicles. For example, the models are divided into flat cars, dump trucks, tractors, box trucks, semi-trailers, etc. The vehicle model matching label is used to filter out vehicles that meet the requirements of the source of goods.

[0059] See also Figure 3 Optionally, the method for recommending a source of goods to a driver provided in an embodiment of the present application includes the following steps:

[0060] Step S301: Driver information collection. The driver information may include the driver's ID number, the driver's name, the driver's license information, the driver's qualifications, the driver's ID information, the driver's place of residence, the routes the driver often travels, the driver's historical score, etc.

[0061] Step S302: Collect basic information of transport vehicles. Transport vehicles are vehicles, ships, and other transport vehicles that have the function of transporting goods. The basic information of transport vehicles includes the transport vehicle license plate (such as license plate number, ship number), transport vehicle type (such as vehicle model, ship type), load capacity, length, width, height, transportation purpose, loading and unloading method, etc.

[0062] Step S303: Collection of existing transport plan information. Existing transport plans refer to existing tasks of the driver, including transport task number, transport tool information, driver information, transport start time, transport end time, loading location (GPS longitude and latitude coordinates), unloading location (GPS longitude and latitude coordinates), task classification (history / ongoing / planned), etc. For example, driver Zhang San already has 2 tasks, task 1 is to depart from Tianjin on January 4, transport a batch of goods, and arrive in Taiyuan on January 8, and task 2 is to depart from Xi'an on January 12, transport a batch of goods, and arrive in Guangzhou on January 17.

[0063] Step S304: Collection of cargo source information and transportation requirements. The cargo source information and transportation requirements may include cargo information, cargo owner information, transportation start time, transportation end time, loading location (GPS latitude and longitude coordinates), unloading location (GPS latitude and longitude coordinates), transportation freight, etc. For example, cargo source 1 needs to depart from Jinan on January 2 and arrive in Tianjin on January 5, and the price the cargo owner is willing to pay is 15 yuan / ton-kilometer. Cargo source 2 needs to depart from Taiyuan on January 7 and arrive in Xi'an on January 14, and the price the cargo owner is willing to pay is 18 yuan / ton-kilometer. Cargo source 3 needs to depart from Jinan on January 1 and arrive in Baotou on January 4, and the price the cargo owner is willing to pay is 20 yuan / ton-kilometer. Cargo source 4 needs to depart from Chengdu on January 8 and arrive in Hangzhou on January 12, and the price the cargo owner is willing to pay is 15 yuan / ton-kilometer. Cargo source 5 needs to depart from Jinzhong on January 9 and arrive in Xianyang on January 11, and the price the cargo owner is willing to pay is 15 yuan / ton-kilometer.

[0064] Step S305: Matching supply and demand of transportation resources.

[0065] The matching of supply and demand of transportation resources can include matching in multiple dimensions, such as matching in the time dimension, matching in the location dimension, matching in the vehicle model, matching in the load, matching in the freight rate, matching in the number of historical line operations, matching in the number of historical cooperation between the two parties, etc. The matching priorities of different dimensions are different. For example, the matching of vehicle model and load are at the first priority. Only when the vehicle model and load are successfully matched, will the matching of other dimensions be continued; if one of the vehicle model and load is not matched successfully, or the vehicle model and load are not matched successfully, this means that the vehicle is not capable of loading and transporting the goods. In this case, the matching of other dimensions is meaningless and no matching of other dimensions will be performed. The cargo source information will be cleared and will not be recommended to the driver.

[0066] The matching of time dimension and location dimension are at the second priority. When the vehicle type and load capacity are matched successfully, the matching of time dimension and location dimension are carried out. The matching of time dimension can be carried out first, and then the matching of location dimension; or the matching of location dimension can be carried out first, and then the matching of time dimension. If both dimensions fail to match or one dimension fails to match, the source of goods cannot be matched successfully with the driver, that is, the source of goods information will be cleared and will not be recommended to the driver. Only when both dimensions are matched successfully, the matching of other dimensions will continue, and the source of goods may be recommended to the driver.

[0067] After obtaining the driver's transportation plan information, the driver's free time information is determined based on the transportation plan information. After obtaining the cargo source information and transportation demand, the loading time and location of the cargo source's transportation demand, as well as the unloading time and location of the transportation demand are determined. The scores of the time matching tag and the location matching tag are determined based on the driver's free time information, the loading time and location of the cargo source's transportation demand, and the unloading time and location of the transportation demand.

[0068] The time matching tag includes two types of time matching tags, among which the first type of time matching tag is used to indicate whether the time interval of the transportation demand of the cargo source is within a certain free time interval (free period) of the driver. If the time interval of the transportation demand of the cargo source is included in a certain free period of the driver, the calculation of the matching degree of the second type of time matching tag will continue; if the time interval of the transportation demand of the cargo source is not included in any free period of the driver, it means that the driver does not have enough free time to transport the cargo source, and the cargo source should not be recommended to the driver. In this case, the calculation of the matching degree of the second type of time matching tag will no longer be performed. The second type of time matching tag is used to indicate the length of the time interval between the two endpoints of the time interval of the transportation demand of the cargo source and the two endpoints of the free time interval of the driver when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver.

[0069] When the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver, the score of the location matching label is calculated, wherein the location matching label indicates the distance between the loading location of the transportation demand of the source of cargo and the driver's location at the left end point of the driver's free time interval, and the distance between the unloading location of the transportation demand of the source of cargo and the driver's location at the right end point of the driver's free time interval.

[0070] The current time is t0, the driver's location is L0, and the driver has 2 transportation tasks. The first transportation task is to depart from location L1 at t1 and arrive at location L2 at t2, and the second transportation task is to depart from location L3 at t3 and arrive at location L4 at t4, among which t0<t1<t2<t3<t4.

[0071] The driver's idle period (idle time interval): (t0, t1), (t2, t3), (t4, t*). t*>t4, which is a certain moment in the future. The left endpoint of the idle time interval is the start time of the idle time interval, and the right endpoint of the idle time interval is the end time of the idle time interval. The value of the left endpoint of the same idle time interval is smaller than the value of the right endpoint. The left endpoint of the idle time interval (t0, t1) is t0; the right endpoint is t1. The left endpoint of the idle time interval (t2, t3) is t2; the right endpoint is t3. The left endpoint of the idle time interval (t4, t*) is t4; the right endpoint is a certain moment in the future.

[0072] The cargo needs to depart from location LA at time tA and arrive at location LB at time tB. Then the time interval of the cargo transportation demand is (tA, tB). Among them, t0≤tA<tB.

[0073] If tB≤t1 or t2≤tA<tB≤t3 or tA≥t4, then the time interval of the transportation demand of the cargo source is included in a certain idle period of the driver, and the first type of time matching tag matching is successful. Otherwise, the time interval of the transportation demand of the cargo source is not included in any idle period of the driver, and the first type of time matching tag matching fails. In the case of successful matching of the first type of time matching tag, the next step of matching is performed; in the case of failed matching of the first type of time matching tag, no further matching is performed, the data of the cargo source is cleared, and the cargo source will not be recommended to the driver. Because the failure of the first type of time matching tag matching means that the transportation demand of the cargo source conflicts with the driver's existing transportation plan in time, and the driver does not have time to transport the cargo source. Even if other conditions are met (matched), the driver cannot undertake the transportation task of the cargo source. Therefore, no longer considering whether other conditions are met (matched) at all, the data of the cargo source can be cleared, and it is determined that the cargo source will not be recommended to the driver, thereby improving the matching efficiency.

[0074] The second type of time matching tag is used to indicate the length of the time interval between the two endpoints of the time interval of the cargo transportation demand and the two endpoints of the driver's free time interval when the time interval of the cargo transportation demand is within a certain free time interval of the driver. Specifically, the shorter the time interval between the left endpoint of the time interval of the cargo transportation demand and the left endpoint of the driver's free time interval, and the shorter the time interval between the right endpoint of the time interval of the cargo transportation demand and the right endpoint of the driver's free time interval, the higher the score of the second type of time matching tag, indicating that taking over the cargo source can make very efficient use of the driver's free time, reduce the idle driving rate, and improve the vehicle's operating efficiency.

[0075] The first case: tB≤t1, the time interval of the cargo transportation demand is included in the driver's idle period (t0, t1), then the first type of time matching tag matching is successful, the driver needs to rush from the current location L0 to the location LA, and the time of arrival at the location LA cannot be later than tA. Then the driver loads the cargo, departs from the location LA at time tA, and arrives at the location LB at time tB, thus completing the transportation of the cargo. The driver unloads the cargo at the location LB at time tB and rushes to the location L1, and the time of arrival at the location L1 cannot be later than t1.

[0076] The two endpoints of the time interval of the transportation demand of the cargo source are tA and tB respectively. The two endpoints of the idle time interval of the driver are t0 and t1 respectively.

[0077] The shorter the time interval between the left endpoint tA of the time interval of the cargo transportation demand and the left endpoint t0 of the driver's idle time interval, the shorter the time interval between the right endpoint tB of the time interval of the cargo transportation demand and the right endpoint t1 of the driver's idle time interval, the higher the score of the second type of time matching label, indicating that taking over the cargo can make very efficient use of the driver's idle time. In terms of time, after transporting the cargo, the first existing transportation task mentioned above can be quickly connected.

[0078] The second case: t2≤tA<tB≤t3, the time interval of the cargo transportation demand is included in the driver's idle period (t2, t3), then the first type of time matching tag matching is successful, the driver needs to rush from location L2 to location LA, and the time of arrival at location LA cannot be later than tA. Then the driver loads the cargo, departs from location LA at time tA, and arrives at location LB at time tB, thus completing the transportation of the cargo. The driver unloads the cargo at location LB at time tB and rushes to location L3, and the time of arrival at location L3 cannot be later than t3.

[0079] The two endpoints of the time interval of the transportation demand of the cargo source are tA and tB respectively. The two endpoints of the idle time interval of the driver are t2 and t3 respectively.

[0080] The shorter the time interval between the left endpoint tA of the time interval of the cargo transportation demand and the left endpoint t2 of the driver's idle time interval, the shorter the time interval between the right endpoint tB of the time interval of the cargo transportation demand and the right endpoint t3 of the driver's idle time interval, the higher the score of the second type of time matching label, indicating that taking over the cargo can make very efficient use of the driver's idle time. In terms of time, after the completion of the first existing transportation task, the transportation of the cargo can be quickly connected, and after transporting the cargo and unloading the cargo, the second existing transportation task can be quickly connected.

[0081] The third case: tA ≥ t4, the time interval of the cargo transportation demand is included in the driver's idle period (t4, t*), then the first type of time matching tag matching is successful, the driver needs to rush from location L4 to location LA, and the time to arrive at location LA cannot be later than tA.

[0082] The two endpoints of the time interval of the transportation demand of the cargo source are tA and tB respectively. The two endpoints of the idle time interval of the driver are t4 and t* respectively.

[0083] The shorter the time interval between the left endpoint tA of the time interval of the cargo transportation demand and the left endpoint t4 of the driver's idle time interval, the higher the score of the second type of time matching label, indicating that taking over the cargo can make very efficient use of the driver's idle time. In terms of time, the transportation of the cargo can be quickly connected after the completion of the second existing transportation task.

[0084] In all three cases, drivers are required to rush from one location to another, which involves the problem of matching the location dimension.

[0085] When the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver, the score of the location matching tag is calculated, wherein the location matching tag indicates the distance between the loading location of the transportation demand of the source of cargo and the location of the driver at the left end point of the driver's free time interval, and the distance between the unloading location of the transportation demand of the source of cargo and the location of the driver at the right end point of the driver's free time interval when the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver. The shorter the distance between the loading location of the transportation demand of the source of cargo and the location of the driver at the left end point of the driver's free time interval, and the shorter the distance between the unloading location of the transportation demand of the source of cargo and the location of the driver at the right end point of the driver's free time interval, the higher the location matching tag score.

[0086] The first case is taken as an example for detailed description below.

[0087] If the distance between the current location L0 and location LA is far, or the distance between location LB and location L1 is far, on the one hand, the driver may not be able to make it due to limited time, and on the other hand, even if he can make it in time, it is not cost-effective in terms of efficiency and economy. Such a source of goods is not an ideal source of goods for the driver. For example, t0 is February 3, L0 is Beijing; tA is February 4, LA is Guangzhou; tB is February 8, LB is Chongqing; t1 is February 10, and L1 is Harbin. The driver is in Beijing on February 3. If he wants to transport this source of goods, he needs to rush to Guangzhou on February 4. However, the distance between Beijing and Guangzhou is very far and cannot be reached within 1 day. This determines that the driver cannot transport this source of goods. Although the time interval of the transportation demand of this source of goods is within the driver's idle period, the distance dimension does not match, and this source of goods should not be recommended to the driver. The ideal situation is that the distance between the current location L0 and location LA is close, and the distance between location LB and location L1 is close. In this way, not only can the driver get there within a limited time, but the time spent on the road is also shorter, which is cost-effective in terms of efficiency and economy. "Close distance" is a relatively general term. How close is the distance? This "close distance" is not necessarily a fixed value. It can be determined by many factors: the time interval between the two end points of the time interval of the cargo transportation demand and the two end points of the driver's idle period, the distance the driver is willing to accept for empty running, the length of time the driver is willing to rest, etc.

[0088] It is mentioned in step S303 that the driver Zhang San already has 2 tasks. Task 1 is to depart from Tianjin on January 4, transport a batch of goods, and arrive in Taiyuan on January 8. Task 2 is to depart from Xi'an on January 12, transport a batch of goods, and arrive in Guangzhou on January 17.

[0089] The cargo source 1 mentioned in step S304 should not be recommended to driver Zhang San. Because if cargo source 1 is recommended to Zhang San, then Zhang San will depart from Jinan on January 2 to transport cargo source 1 and arrive in Tianjin on January 5, while Zhang San's original task 1 needs to depart from Tianjin on January 4, which delays task 1.

[0090] The cargo source 2 mentioned in step S304 should not be recommended to Zhang San either, because the cargo source 2 needs to depart from Taiyuan on January 7, while Zhang San is still on the way from Tianjin to Taiyuan on January 7, and can only arrive in Taiyuan on January 8.

[0091] The cargo source 3 mentioned in step S304 should not be recommended to Zhang San. Because if cargo source 3 is recommended to Zhang San, Zhang San will depart from Jinan on January 1 to transport cargo source 1 and arrive in Baotou on January 4. However, Zhang San's original task 1 needs to depart from Tianjin on January 4. Baotou is far away from Tianjin. After Zhang San arrives in Baotou on January 4, he cannot depart from Tianjin to perform task 1 on the same day, which delays task 1.

[0092] The cargo source 4 mentioned in step S304 should not be recommended to Zhang San either, because cargo source 4 needs to depart from Chengdu on January 8, and Zhang San arrives in Taiyuan after completing task 1 on January 8. Taiyuan is far away from Chengdu, so he cannot rush to Chengdu to transport cargo source 4 on the same day.

[0093] The cargo source 5 mentioned in step S304 can be recommended to Zhang San. The cargo source 5 needs to depart from Jinzhong on January 9 and arrive in Xianyang on January 11. After Zhang San completes Task 1, he is in Taiyuan on January 8. Since Taiyuan is very close to Jinzhong, Zhang San can rush to Jinzhong before January 9, load cargo source 5 in Jinzhong and depart on January 9, arrive in Xianyang on January 11, unload cargo source 5, and rush to Xi'an. Since Xianyang and Xi'an are very close, Zhang San can arrive in Xi'an before January 12 to perform Task 2.

[0094] The cargo sources 1, 2, 3, and 4 mentioned in step S304 all failed to match with the driver Zhang San, and only the cargo source 5 successfully matched with Zhang San.

[0095] Step S306: Enter the matching recommendation process again, and no longer match the failed parties in the short term. In the short term, cargo sources 1, 2, 3, 4 and driver Zhang San will no longer be matched.

[0096] Step S307: insert matching transport tasks and update the transport plan, and match again according to the new transport plan. Update Zhang San's transport plan according to the transport demand of source 5, and use Zhang San's updated transport plan for matching next time.

[0097] The method provided in this application obtains information such as the idle time, location, in-transit and idle status of transportation resources based on short- and long-term transportation capacity plans, accurately matches the start and end time, loading and unloading locations and other information of the shipment plan, and combines the planned transportation demand on the transportation capacity demand side with the transportation plan of social transportation capacity resources, so that the two can be matched more efficiently and accurately, playing a role in preparing for a rainy day, planning shipment, planning transportation, and continuous and efficient operation.

[0098] As an optional implementation, after step S304 and before step S305, the supply is spliced, and step S305 is performed on the spliced ​​supply.

[0099] The specific method of splicing the cargo sources is: obtaining matching cargo sources that match the "transportation tool information" of the driver; splicing matching cargo sources that are continuous in time and location to form a spliced ​​cargo source, and adding the spliced ​​cargo source to the cargo source information. The continuity of two cargo sources in time means that the time interval between the unloading (unloading) time of one cargo source and the loading (loading) time of another cargo source is within a preset time interval. The continuity of two cargo sources in location means that the distance between the unloading (unloading) location of one cargo source and the loading (loading) location of another cargo source is within a preset distance.

[0100] For example, the transportation demand of source 1 starts on January 3, the loading location is A, the end time is January 8, and the unloading location is B; the transportation demand of source 2 starts on January 8, the loading location is C, the end time is January 12, and the unloading location is D. If the distance between A and B is within the preset distance, then source 1 and source 2 are spliced ​​to obtain new source information: source 1'. The transportation demand of source 1' starts on January 3, the loading location is A, the end time of the transportation demand is January 12, and the unloading location is D.

[0101] For another example, the transportation demand of source 3 starts on January 2, the loading location is E, the end time of the transportation demand is January 6, and the unloading location is F; the transportation demand of source 4 starts on January 7, the loading location is G, the end time is January 9, and the unloading location is H; the transportation demand of source 5 starts on January 10, the loading location is I, the end time is January 15, and the unloading location is J. If the distance between F and G is within the preset distance, and the distance between H and I is within the preset distance, then sources 3, 4, and 5 are concatenated to obtain new source information: source 2'. The transportation demand of source 2' starts on January 2, the loading location is E, the end time of the transportation demand is January 15, and the unloading location is J.

[0102] After the step of cargo source splicing is completed, the cargo source information obtained after splicing is used to match the driver, and the matching includes matching in the time dimension, matching in the location dimension, etc. The advantages of splicing the cargo sources first and then matching the spliced ​​cargo sources with the driver are: two or more cargo sources can be matched at one time, which improves the matching efficiency. When recommending cargo sources to drivers, two or more cargo sources that are closely connected in time and location can be recommended at one time, which improves the efficiency of recommending cargo sources to drivers.

[0103] like Figure 4 As shown, the embodiment of the present application includes the following steps:

[0104] Step S401: Obtain driver and vehicle information, obtain known transportation plans of the driver and vehicle, and obtain cargo source information.

[0105] Driver information may include the driver's ID number, driver's name, driver's license information, driver's qualifications, driver's ID information, driver's place of residence, routes the driver frequently travels, driver's historical scores, etc.

[0106] Vehicle information includes license plate number, vehicle model, load capacity, length, width and height, transportation purpose, loading and unloading methods, etc.

[0107] The existing transport plan refers to the existing tasks of the driver, including the transport task number, transport tool information, driver information, transport start time, transport end time, loading location (GPS longitude and latitude coordinates), unloading location (GPS longitude and latitude coordinates), task classification (history / ongoing / planned), etc. For example, driver Zhang San already has two tasks, task 1 is to depart from Tianjin on January 4, transport a batch of goods, and arrive in Taiyuan on January 8, and task 2 is to depart from Xi'an on January 12, transport a batch of goods, and arrive in Guangzhou on January 17.

[0108] The cargo source information and transportation requirements may include cargo information, cargo owner information, transportation start time, transportation end time, loading location (GPS longitude and latitude coordinates), unloading location (GPS longitude and latitude coordinates), transportation freight, etc. For example, cargo source 1 needs to depart from Jinan on January 2 and arrive in Tianjin on January 5. Cargo source 2 needs to depart from Taiyuan on January 7 and arrive in Xi'an on January 14. Cargo source 3 needs to depart from Jinan on January 1 and arrive in Baotou on January 4. Cargo source 4 needs to depart from Chengdu on January 8 and arrive in Hangzhou on January 12. Cargo source 5 needs to depart from Jinzhong on January 9 and arrive in Xianyang on January 11.

[0109] Step S402: Determine the driver's designated transportation idle period (idle period).

[0110] The designated transportation gap of the driver can be determined by the known transportation plan of the driver. For example, assuming that the current time is January 1, the driver Zhang San has 2 tasks, task 1 is to depart from Tianjin on January 4, transport a batch of goods, and arrive in Taiyuan on January 8, and task 2 is to depart from Xi'an on January 12, transport a batch of goods, and arrive in Guangzhou on January 17, then Zhang San's designated transportation gap period is January 1 to January 4, January 8 to January 12, and January 17 to a future time.

[0111] Step S403: Clean all the cargo information data. Specifically, determine whether the transportation range of the cargo information is included in the start time to the end time of the specified transportation gap period. If yes, it will be retained; if not, it will be removed from the object sample pool.

[0112] Source 1 needs to depart from Jinan on January 2 and arrive in Tianjin on January 5. Source 2 needs to depart from Taiyuan on January 7 and arrive in Xi'an on January 14. Source 3 needs to depart from Jinan on January 1 and arrive in Baotou on January 4. Source 4 needs to depart from Chengdu on January 8 and arrive in Hangzhou on January 12. Source 5 needs to depart from Jinzhong on January 9 and arrive in Xianyang on January 11.

[0113] The transportation interval of source 1 is (January 2, January 5), which is not included in any of the three specified transportation gaps mentioned above. Therefore, source 1 belongs to the failed matching source and its data is cleared.

[0114] The transportation interval of source 2 is (January 7, January 14), which is not included in any of the three specified transportation gaps mentioned above. Therefore, source 2 belongs to the failed matching source and its data is cleared.

[0115] The transportation interval of source 3 is (January 1, January 4), which is included in the specified transportation gap period (January 1, January 4).

[0116] The transportation interval of source 4 is (January 8, January 12), which is included in the specified transportation gap period (January 8, January 12).

[0117] The transportation interval of source 5 is (January 9, January 11), which is included in the designated transportation gap period (January 8, January 12).

[0118] The data of sources 1 and 2 are cleared, while the data of sources 3, 4, and 5 are retained.

[0119] Step S404: Perform factor matching scoring and combined data modeling on the cleaned data, and output the recommended results of the supply source information arrangement and combination.

[0120] Factor matching can include matching in multiple dimensions, such as location matching, freight matching, historical route operation times matching, historical cooperation times matching, etc. Different factors have different weights when matching. The more important the factor, the higher the weight. The matching degree between the source of goods and the driver is calculated based on the scores and weights of multiple factors.

[0121] This recommendation algorithm will be based on decision tree and decision weighted scoring machine learning. The main steps include: sorting out non-leaf node elements based on decision tree; weighted scoring based on leaf node categories of decision tree; weighted setting and machine learning scheme under cold start.

[0122] Step S405: The output cargo source information arrangement and combination recommendation results are selected by the driver, and the matching success information is recorded, and a new transportation task plan after matching is generated.

[0123] Step S406: The specific schedule recommendation is completed.

[0124] The present application relates to a supply and demand matching method based on the task idle periods in the short- and long-term transportation task plans of the capacity provider (driver) and the short- and long-term transportation capacity demands of the capacity demander (the cargo owner). The method uses a decision tree to classify the elements of the matching dimension, improves the decision-making mechanism, verifies and corrects the results through machine learning, and ultimately achieves accurate matching of supply and demand of capacity resources and improves the planning of short- and long-term transportation tasks.

[0125] As an optional implementation, the method for recommending cargo sources to drivers provided in the present application may include two parts: first, weighted score recommendation; second, machine learning weight optimization. These two parts are described in detail below.

[0126] 1. Weighted score recommendation:

[0127] Definition 1. Recommended tags

[0128] Recommended tags are conditional dimensions used to calculate matching. They are defined as:

[0129] t=(t 1 , t 2 ,…,t p )

[0130] where t k The label is the kth conditional dimension. Subsequent weighting is also set based on this dimension.

[0131] Definition 2. Label weight

[0132] Tag weight refers to the weight value of the recommended tag.

[0133] s=(t,tagRating)

[0134] Among them, t is the recommended tag, and tagRating is the weight attribute. The more important the dimension, the higher the absolute value of the weight, and vice versa.

[0135]

[0136] Notes:

[0137] Load compliance (※1) refers to the cargo owner's requirements for the carrying capacity of the transport vehicle, including the matching degree between the shipment weight and the carrying vehicle's load capacity, or the matching degree between the cargo volume and the space of the transport vehicle.

[0138] Time compliance (※2) refers to whether the transportation time interval required by the shipper, i.e. the transportation start and end time periods, is within the idle time interval of the driver's current or future transportation tasks. In other words, it does not conflict with the existing task plan time.

[0139] The distance between the departure point of the transport task and the idle period start point (※3) refers to the distance between the loading point specified in the transport order of the shipper and the current or future idle start point of the driver. The idle start point refers to the current location of the driver if the driver is currently idle (no transport task); if the driver is idle for a certain period of time in the future, it is the unloading location of the nearest transport task before the future idle time. The closer the distance between the shipper's loading point and the driver's idle start point is, the higher the score.

[0140] The short time interval between the departure time of the transport task and the start of the idle period (※4) refers to the time interval between the loading time specified in the transport request of the shipper and the current or future idle start time of the driver. The idle start time refers to the current time if the driver is currently idle (no transport task); if the driver is idle for a certain period of time in the future, it refers to the end time of the most recent transport task before the future idle time. The shorter the time interval between the shipper's delivery time and the driver's idle start time, the higher the score.

[0141] The distance between the arrival point of the transport task and the end point of the idle period (※5) refers to the distance between the unloading point specified in the transport commission of the cargo owner and the current or future idle end point of the driver. The idle end point refers to the departure point of the nearest transport task after now if the driver is currently idle (no transport task); if the driver is idle for a certain period of time in the future, it is the departure point of the nearest transport task after the idle time in the future. If there is no transport task, this dimension is not scored. The closer the distance between the cargo owner's unloading point and the driver's idle end point is, the higher the score.

[0142] The short time interval between the arrival time of the transport task and the end of the idle period (※6) refers to the interval between the unloading time specified in the transport commission of the cargo owner and the current or future idle end time of the driver. The idle end time refers to the departure time of the nearest transport task after now if the driver is currently idle (no transport task); if the driver is idle for a certain period of time in the future, it is the departure time of the nearest transport task after the future idle time. If there is no transport task, this dimension is not scored. The shorter the interval between the cargo owner's unloading time and the driver's idle end time, the higher the score.

[0143] The freight unit price (※7) refers to the degree of price matching between the shipper and the driver. When recommending from the shipper's perspective, the lower the freight price, the higher the score; when recommending from the driver's perspective, the higher the freight price, the higher the score.

[0144] Figure 2-1 ,2-2 , 2-3, and 2-4 show the matching of different cargo sources and drivers. The detailed description is as above and will not be repeated here.

[0145] Definition 3. Data set based on label weights

[0146] In the recommendation algorithm based on tag weight scoring, the data representation can be described as a set of five-valued data.

[0147] M: = (U, I, R, S, Y)

[0148] U is a user (driver) task information set, which can also be understood as a data set based on the driver's current and future task plans. where u k (k=1, 2, ..., m) is the kth basic attribute of the user.

[0149] Combination Figures 2-1 to 2-4 , assuming that the existing information of a user "Zhang San" is as follows:

[0150] {5047,"zhangsan","20210101","idle","Jinan",("Jin A34567",35,"3"),[(20210104,"Tianjin","start"),(20210108,"Taiyuan","end")][(20210112,"Xi'an","start"),(20210117,"Guangzhou","end")],…}

[0151] This information means: Assuming today is January 1, 2020, user Zhang San, ID 5047, is currently idle and located in Jinan, Shandong. In the future, he will drive the vehicle model 3 (cangzha car) with a load capacity of 35 tons, Tianjin A34567, departing from Tianjin on January 4, 2021, and deliver goods to Taiyuan, Shanxi before January 8, 2021. Depart from Xi'an, Shaanxi on January 12, 2021, and arrive in Guangzhou before January 17, 2021. Subsequent other transportation plans are replaced by "..."

[0152] In subsequent calculations, the algorithm will make corresponding supply recommendations for the user's free time. This process is a cyclic operation, so the information of user Zhang San will be split as follows.

[0153] U 1 ={5047,"zhangsan",("Jin A34567",35,"3"),[(20210104,"Tianjin","start"),(20210108,"Taiyuan","end")]}

[0154] U 2={5047,"zhangsan",("Jin A34567",35,"3"),[(20210112,"Xi'an","start"),(20210117,"Guangzhou","end"]}

[0155] …

[0156] According to the above disassembly example, ignoring some basic properties can understand U 1 The basic properties of u 1 For loading time 20210104, u 2 The loading location is Tianjin, etc.

[0157] Through the above splitting, the supply information will be recommended mainly based on Zhang San’s free time. For example, Zhang San’s first free time information u n’ As shown below, this step is the internal information conversion process of the algorithm and is only for understanding.

[0158] u 1’ ={5047,"zhangsan",("Jinan A34567",35,"3"),[(20210101,"Jinan","end"),(20210104,"Tianjin","start")]}

[0159] u 2’ ={5047,"zhangsan",("Jin A34567",35,"3"),[(20210108,"Taiyuan","end"),(20210112,"Xi'an","start")]}

[0160] I is the source information set, where i k (k=1, 2, ..., m) is the kth basic attribute of the source of goods. For the convenience of explanation, the source of goods information takes a single task as an example, assuming that the information released by a certain owner is as follows:

[0161] i={30,3,"Sand and gravel",(20210109,"Jinzhong",start),(20210111,"Xianyang",end)}

[0162] It means that the cargo volume of sand and gravel is 30 tons, and the required vehicle type number 3 (cangzha truck) is required. It will depart from Jinzhong, Shanxi on January 9, 2021, and arrive in Xianyang, Shaanxi on January 11, 2021.

[0163] According to the above disassembly example, ignoring some basic attributes can understand the basic attribute i 1 For loading time 20210109, i 2 The loading location is Jinzhong, etc.

[0164] R represents the set of drivers’ preference scores for cargo source information, where r k represents the preference score of user u for the kth basic attribute of source i. For example, in the above example, Zhang San and the source information have a preference of 1 day for loading time, that is, Zhang San u 2’ The preference between (20210108, "Taiyuan", "end") and (20210109, "Jinzhong", start) in source i, i = (1, ..., ..., ...), which means that the preference score between the driver and the delivery time of the source of cargo is 1 point.

[0165] S is the user's label weight set for the source information. Where t is the tag attribute and tagRating is the weight attribute. And it satisfies Where T is the user's label set for the source of goods information.

[0166] Y is the four-element relationship between U, I, R, and S, and satisfies For example, y = (u, i, r, (t, tagRating)) ∈ Y indicates that user u's preference score for source information i is r, and the weight of the first tag of source information i is specified as s = (t, tagRating). Among them, t is one of the features that user u believes source information i has, and the weight of the tag is tagRating.

[0167] Definition 4. Related symbols

[0168] (1) s(u, i)∈S represents the weighted score of one of the labels of the source information i given by user u;

[0169] (2) Represents a set of label weight scores of user u on the source information i;

[0170] (3) S(U, I) = {s(u, i|u∈U, i∈I)} represents the set of label weight scores of users in the user set U for the supply information in the supply information set I;

[0171] (4)s(u)∈S represents one of the label weight features of user u;

[0172] (5) Represents the tag weight feature set of user u;

[0173] (6) S(U) = {s(u)|u∈U} represents the tag weight feature set of users in the user set U;

[0174] (7)s(i)∈S represents one of the label weight features of the source information i;

[0175] (8) Represents the label weight feature set of the source information i;

[0176] (9) S(I) = {s(i)|i∈I} represents the label weight feature set of the supply information in the supply information set I;

[0177] (10) r(u, i)∈R represents the preference score of user u for the source information i;

[0178] (11) r(i)∈R represents the overall score of the source information i;

[0179] (12) R(I) = {r(i)} represents the overall score of the supply information set I;

[0180] (13) R(U, I) = {r(u, i|u∈U, i∈I)} represents the preference rating set of users in the user set U for the supply information in the supply information set I.

[0181] Definition 5. Algorithm Flow

[0182] Input: Rating data D = {(u 1 ,i 1 , r 1 ,s 1 ), (u 2 ,i 2 , r 2 ,s 2 ),…,(u N ,i N , r N ,s N )}

[0183] Output: Label weight feature set S(I) = {s(i)|i∈I} of the source information

[0184] Algorithm process:

[0185] 1, I = {i|(u, i, r, s)∈D} / / Objects participating in label weight calculation

[0186] 2, S(I) = {S(i)}, / / The initial label weight feature is empty

[0187] 3.foreach(u,i,r,s)∈D do

[0188] 4. t=getTag(s); tagRating=getTagRating(s);

[0189] 5. addToSlider(S(i), t, r, tagRating); / / Increase the weight of the corresponding feature

[0190] 6. normalize(S(i))

[0191] 7, return S(I)

[0192] The algorithm flow is as follows: First, the calculated source information tag weight feature is initialized to empty (lines 1 to 2), and for any tuple in D, a loop judgment is performed (lines 3 to 5). First, the tag feature t of the source information i and the weight tagRating of the tag feature are separated (line 4), and the current feature weight is cumulatively added to the corresponding tag weight feature S(i) of the source information i (line 5). Each tuple is judged iteratively until all tuples are judged (lines 3 to 5), and finally S(i) is normalized (line 6), and the tag weight feature S(i) of all items involved in the tag weight calculation is returned (line 7).

[0193] A brief operation example:

[0194] Based on the delivery time difference (hours) and the number of historical cooperation (times), the preference results are:

[0195] i=(24h,…,15,…)

[0196] If the label weight is

[0197] s = (240, ..., 2.5, ...)

[0198] The simplified weighted calculation is

[0199] S(i)=(240 / 24,…,2.5×15,…)

[0200] Assuming that the delivery time difference is 24 hours and the preset weight is 240 hours, the weight score based on the time difference is 240 / 24 = 10 points. When the number of successful cooperation in history is 15 times and the preset weight is 2.5 times, the weight score based on cooperation preference is 2.5X15 = 37.5 points. The lower the preference, the higher the recommendation, the lower limit of the label is defined. For example, the minimum time difference is defined as 4 hours, and less than 4 hours is considered as 4 hours. The higher the preference, the higher the recommendation, and the upper limit of the score is set according to the situation.

[0201] 2. Machine Learning Weight Optimization:

[0202] The algorithm analyzes the priority recommendation results through user feedback and optimizes the weight shares (weight values) in the algorithm.

[0203] Definition 1. Scoring Indicators

[0204] For the accuracy and optimization of recommendation results, the indicators for measuring recommendation quality are also different. The following two indicators are used as the basis for measuring the accuracy of the scoring algorithm:

[0205]

[0206] Among them, R represents the user's rating matrix for the recommended source information, R i，j R^ represents the score (feedback result) of the i-th user for the j-th source of goods information, i，j Represents the score predicted by the recommendation algorithm (recommendation result). Ω represents the set of observed scores, and N is the number of observed scores. For RMSE, the smaller the value, the higher the accuracy of the recommendation. For example, the recommended user's source information is ranked by recommendation, n = (1, 2, 3, ...), and the user did not choose the recommendation of n = 1, but chose the recommendation of n = 3, then the recommendation score of n = (1, 2) should be between n = (3, ..., N).

[0207] Definition 2. Machine Learning Correction and Regression Testing

[0208] If R i，j Feedback result items and R^ i，j When recommendation results have different label scores, machine learning is used to correct them.

[0209]

[0210] As recommended above:

[0211] y=(u,i,r,(t,tagRating))∈Y

[0212] For the correction weight tR i Correction method:

[0213] E(TR i，j )={e(tR i，j |i,j∈[α,β])}

[0214] or:

[0215]

[0216] where tr i，j Represents the weight tagRating of the correction object, α, β represent the horizontal and vertical coordinate sets of the matrix where the basic attributes of the correction item are located, E(TR i，j ) represents the correction formula that makes y=(u, i, r, (t, e(tagRating)))∈Y result r have the smallest RMSE value among the recommended selection results.

[0217] Definition 3. Algorithm Flow

[0218] Input: Correction data D = {tr 1，1 , tr 1，2 ,…,tr i，j}

[0219] Output: The label weight result set E(TR i，j )={e(tR i，j |i,j∈[α,β])}

[0220] Algorithm process:

[0221] 1.TR i，j ={i, j∈[α, β]} / / Objects participating in label weight calculation

[0222] 2, / / The initial label weight feature is empty

[0223] 3. foreach(i,j)∈[α,β]do

[0224] 4. E(TR i，j )=editTag(tr 1，1 , tr 1，2 ,…,tr i，j );

[0225] 5. setTagRating(e, t); / / reset the weight of the corresponding feature

[0226] 6. normalize(E(tr i，j ))

[0227] 7, return E (TR)

[0228] The algorithm flow is as follows: First, the corrected cargo source information label weight feature is initialized to empty (lines 1 to 2), and for any tuple in D, a loop judgment is performed (lines 3 to 5). First, the label weight tr of the cargo source information set is substituted, and the adjusted feature weight is reset and added to the corresponding label weight feature t of the driver (line 5). Each tuple of the reset object is iterated until all tuples are judged (lines 3 to 5), and finally the weight E(tr) is corrected (line 6), and the label weight feature E(TR) of all items involved in the label weight calculation is returned (line 7).

[0229] For example, there are 100 sources of goods to be matched, namely source 1, source 2, ..., source 100, and there are 12 tags. Match source 1 with driver Zhang San to obtain the score of each of the 12 tags, and calculate the matching score S1 between source 1 and Zhang San according to the score of each of the 12 tags and the weight of each tag. Match source 2 with driver Zhang San to obtain the score of each of the 12 tags, and calculate the matching score S2 between source 2 and Zhang San according to the score of each of the 12 tags and the weight of each tag. ... Match source 100 with driver Zhang San to obtain the score of each of the 12 tags, and calculate the matching score S100 between source 100 and Zhang San according to the score of each of the 12 tags and the weight of each tag.

[0230] In the above recommendation algorithm, the selection of the weight value of the tag is very important. When the weight value of the tag changes, the recommendation result often changes accordingly. The method for recommending cargo sources to drivers provided in this application can optimize the weight value of the tag, thereby achieving the best recommendation effect.

[0231] Assume that among the above 100 sources of goods, the matching scores are ranked in order of source 91, source 27, source 11, source 56, source 82... The source information of the first 10 of these 100 sources of goods is displayed and recommended to Zhang San. Zhang San did not choose source 91 with the highest matching score, but chose source 11 with the third highest matching score. This shows that the calculated matching ranking is not exactly equal to the matching ranking in the mind of the user (driver). The matching ranking calculated by the method for recommending sources of goods to drivers provided in this application shows that the matching score of source 11 is lower than the matching score of source 91 and lower than the matching score of source 27. However, the driver believes that source 11 is more matched than sources 91 and 27. This requires us to adjust the method for recommending cargo sources to drivers provided in this application. Specifically, adjust the weight of the label, recalculate the matching score between each cargo source and the driver based on the score of each label and its adjusted weight, and re-recommend the cargo source based on the matching score; obtain the driver's score for the re-recommended cargo source; calculate the recommended deviation parameter based on the deviation between the matching score between the cargo source and the driver and the driver's score for the re-recommended cargo source; determine the weight of the label when the recommended deviation parameter is the smallest, as the weight of the optimized label. The deviation parameter can be the above-mentioned RMSE.

[0232] Assume that a total of 5,000 cargo sources and 200 drivers are used to optimize the method of recommending cargo sources to drivers provided in this application.

[0233] The matching scores of each of the 5,000 cargo sources and each of the 200 drivers are calculated, resulting in a total of 1 million matching scores.

[0234] For driver No. 1, select the top 10 matching scores from 5000 matching scores. Assume that the scores of the top 10 cargo sources are R^ 1，1 , R^ 1，2 , R^ 1，3 , R^ 1，4 , R^ 1，5 , R^ 1，6 , R^ 1，7 , R^ 1，8 , R^ 1，9 , R^ 1，10 The corresponding cargo source information is displayed to driver No. 1 in the order of the scores. Driver No. 1 scores the 10 cargo sources, and the scores are R 1，1 , R 1，2 , R 1，3 , R 1，4 , R 1，5 , R 1，6 , R 1，7 , R 1，8 , R 1，9 , R 1，10 .

[0235] R 1，1 , R^ 1，1 Corresponding to the same source; R 1，2 , R^ 1，2 Corresponding to the same source; R 1，3 , R^ 1，3 Corresponding to the same source; R 1，4 , R^ 1，4 Corresponding to the same source; R 1，5 , R^ 1，5 Corresponding to the same source; R 1，6 , R^ 1，6 Corresponding to the same source; R 1，7 , R^ 1，7 Corresponding to the same source; R 1，8 , R^ 1，8 Corresponding to the same source; R 1，9 , R^ 1，9 Corresponding to the same source; R 1，10 , R^ 1，10 Corresponding to the same source of goods.

[0236] For driver No. 2, select the top 10 matching scores from 5000 matching scores. Assume that the scores of the top 10 cargo sources are R^ 2，1 , R^ 2，2 , R^ 2，3 , R^ 2，4 , R^ 2，5 , R^2，6 , R^ 2，7 , R^ 2，8 , R^ 2，9 , R^ 2，10 , the corresponding cargo source information is displayed to driver No. 2 in the order of the scores, and driver No. 2 scores the 10 cargo sources, with scores of R 2，1 , R 2，2 , R 2，3 , R 2，4 , R 2，5 , R 2，6 , R 2，7 , R 2，8 , R 2，9 , R 2，10 .

[0237] R 2，1 , R^ 2，1 Corresponding to the same source; R 2，2 , R^ 2，2 Corresponding to the same source; R 2，3 , R^ 2，3 Corresponding to the same source; R 2，4 , R^ 2，4 Corresponding to the same source; R 2，5 , R^ 2，5 Corresponding to the same source; R 2，6 , R^ 2，6 Corresponding to the same source; R 2，7 , R^ 2，7 Corresponding to the same source; R 2，8 , R^ 2，8 Corresponding to the same source; R 2，9 , R^ 2，9 Corresponding to the same source; R 2，10 , R^ 2，10 Corresponding to the same source of goods.

[0238] …

[0239] For driver No. 200, select the top 10 matching scores from 5000 matching scores. Assume that the scores of the top 10 cargo sources are R^ 200，1 , R^ 200，2 , R^ 200，3 , R^ 200，4 , R^ 200，5 , R^ 200，6 , R^ 200 , 7 , R^ 200，8 , R^ 200，9 , R^ 200，10, the corresponding cargo source information is displayed to driver No. 200 in the order of the scores. Driver No. 200 scores the 10 cargo sources, and the scores are R 200，1 , R 200，2 , R 200，3 , R 200，4 , R 200，5 , R 200，6 , R 200，7 , R 200，8 , R 200，9 , R 200，10 .

[0240] R 200，1 , R^ 200，1 Corresponding to the same source; R 200，2 , R^ 200，2 Corresponding to the same source; R 200，3 , R^ 200，3 Corresponding to the same source; R 200，4 , R^ 200，4 Corresponding to the same source; R 200，5 , R^ 200，5 Corresponding to the same source; R 200，6 , R^ 200，6 Corresponding to the same source; R 200，7 , R^ 200，7 Corresponding to the same source; R 200，8 , R^ 200，8 Corresponding to the same source; R 200，9 , R^ 200，9 Corresponding to the same source; R 200，10 , R^ 200，10 Corresponding to the same source of goods.

[0241] RMSE = {[(R 1，1 -R^ 1，1 ) 2 +(R 1，2 -R^ 1，2 ) 2 +(R 1，3 -R^ 1，3 ) 2 +(R 1，4 -R^ 1，4 ) 2 +(R 1，5 -R^ 1，5 ) 2 +(R 1，6 -R^ 1，6 ) 2 +(R 1，7 -R^ 1，7 ) 2 +(R 1，8 -R^ 1，8 )2 +(R 1，9 -R^ 1，9 ) 2 +(R 1，10 -R^ 1，10 ) 2 +(R 2，1 -R^ 2，1 ) 2 +(R 2，2 -R^ 2，2 ) 2 +(R 2，3 -R^ 2，3 ) 2 +(R 2，4 -R^ 2，4 ) 2 +(R 2，5 -R^ 2，5 ) 2+(R 2，6 -R^ 2，6 ) 2 +(R 2，7 -R^ 2，7 ) 2 +(R 2，8 -R^ 2，8 ) 2 +(R 2，9 -R^ 2，9 ) 2 +(R 2，10 -R^ 2，10 ) 2 + ……+(R 200，1 -R^ 200，1 ) 2 +(R 200，2 -R^ 200，2 ) 2 +(R 200，3 -R^ 200，3 ) 2 +(R 200，4 -R^ 200，4 ) 2 +(R 200 , 5 -R^ 200，5 ) 2 +(R 200，6 -R^ 200，6 ) 2 +(R 200，7 -R^ 200，7 ) 2 +(R 200，8 -R^ 200，8 ) 2 +(R 200，9 -R^ 200，9 ) 2+(R 200，10 -R^200，10 ) 2 ] / 2000} 1 / 2

[0242] The calculated RMSE represents the deviation of the recommendation. The smaller the value, the higher the accuracy of the recommendation. For a specific set of weight values ​​of the label, the calculated RMSE is unique. When the weight value of the label changes, the matching score between the source of cargo and the driver changes, and the recommendation order also changes. Adjust the weight value of the label, and repeat the above matching, recommendation and RMSE calculation process according to the adjusted weight value of the label; continue to adjust the weight value of the label, and continue to repeat the above matching, recommendation and RMSE calculation process according to the adjusted weight value of the label; until a minimum RMSE value is obtained, then the weight of the label corresponding to this minimum value is the optimized weight.

[0243] like Figure 5 As shown, an embodiment of the present application provides a system for recommending sources of goods to drivers, and the system is used to execute the above-mentioned method for recommending sources of goods to drivers. The system has a user interface, on which the driver can input information about himself and his means of transport, such as driver's license information, vehicle model, vehicle load, frequently traveled routes, and other information. The recommendation engine matches the driver information and the source of goods information according to the driver information and the source of goods information. The matching process includes scoring multiple tags, calculating the matching degree between the source of goods and the driver according to the scores and weights of the tags, ranking according to the matching degree, generating recommendation results, and displaying the recommendation results on the user interface. The driver can choose the source of goods recommended to him on the user interface, and can also score the recommended source of goods. Machine learning is performed based on the driver's score for the recommended source of goods and the system's score for the source of goods, and the weight of the tag is optimized to obtain a better recommendation effect.

[0244] like Figure 6 As shown, an embodiment of the present invention further provides a device for recommending a source of goods to a driver, the device comprising: a first acquisition unit 61 , a second acquisition unit 62 , a first determination unit 63 , a second determination unit 64 , a calculation unit 65 , and a recommendation unit 66 .

[0245] The first acquisition unit 61 is used to acquire the supply information and transportation requirements of multiple supply sources.

[0246] The second acquisition unit 62 is used to acquire driver information and transportation tool information, where the driver information includes basic driver information and transportation plan information.

[0247] The first determination unit 63 is used to determine multiple tags and their weights. The multiple tags include at least a time matching tag and a location matching tag. The time matching tag is used to indicate the matching degree between the driver and the source of cargo in the dimension of time. The location matching tag is used to indicate the matching degree between the driver and the source of cargo in the dimension of location.

[0248] The second determining unit 64 is used to determine the score of each tag in the plurality of tags according to the basic information of the driver, the transportation plan information, the transportation tool information, the cargo source information and the transportation demand.

[0249] The calculation unit 65 is used to calculate the matching score between each cargo source and the driver according to the score of each tag and its weight.

[0250] The recommendation unit 66 is used to recommend cargo sources to the driver according to the matching scores.

[0251] Optionally, the multiple labels also include at least one of the following: a transport model matching label, a transport load matching label.

[0252] Optionally, after obtaining the driver's transportation plan information, the driver's free information is determined based on the transportation plan information, and after obtaining the cargo source information and transportation demand, the second determination unit 64 determines the loading time and place of the cargo source's transportation demand, and the unloading time and place of the transportation demand, and determines the scores of the time matching tag and the place matching tag based on the driver's free information, the loading time and place of the cargo source's transportation demand, and the unloading time and place of the transportation demand.

[0253] Optionally, the time matching tag includes two types of time matching tags, wherein the first type of time matching tag is used to indicate whether the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, and the second type of time matching tag is used to indicate when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, the length of the time interval between the two endpoints of the time interval of the transportation demand of the cargo source and the two endpoints of the free time interval of the driver.

[0254] Optionally, when the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver, the second determination unit 64 calculates the score of the location matching tag, wherein the location matching tag is used to indicate the distance between the loading location of the transportation demand of the source of cargo and the driver's location at the left end point of the driver's free time interval, and the distance between the unloading location of the transportation demand of the source of cargo and the driver's location at the right end point of the driver's free time interval when the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver.

[0255] Optionally, the device further includes an optimization unit for optimizing the weights of each tag, and the optimization unit specifically includes: a recording subunit and an optimization subunit.

[0256] The recording subunit is used to record the driver's evaluation information on the source of goods after recommending the source of goods to the driver according to the degree of matching.

[0257] The optimization subunit is used to optimize the weight of each label according to the driver's evaluation information on the source of goods.

[0258] Optionally, the optimization subunit specifically includes: an adjustment module, an acquisition module, a calculation module, and a determination module.

[0259] The adjustment module is used to adjust the weight of the label, recalculate the matching score between each cargo source and the driver according to the score of each label and its adjusted weight, and re-recommend the cargo source according to the matching score.

[0260] The acquisition module is used to obtain the driver's rating of the re-recommended cargo source.

[0261] The calculation module is used to calculate the recommended deviation parameter according to the deviation between the matching score between the cargo source and the driver and the driver's score for the re-recommended cargo source.

[0262] The determination module is used to determine the weight of the label when the recommended deviation parameter is the smallest, which is used as the weight of the optimized label.

[0263] Figure 7 It is a block diagram of an electronic device according to an exemplary embodiment.

[0264] Refer to the following Figure 7 An electronic device 700 according to this embodiment of the present disclosure is described. Figure 7 The electronic device 700 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0265] like Figure 7 As shown, the electronic device 700 is in the form of a general computing device. The components of the electronic device 700 may include, but are not limited to: at least one processing unit 710, at least one storage unit 720, a bus 730 connecting different system components (including the storage unit 720 and the processing unit 710), a display unit 740, etc.

[0266] The storage unit stores program codes, and the program codes can be executed by the processing unit 710, so that the processing unit 710 executes the steps described in this specification according to various exemplary embodiments of the present disclosure.

[0267] The storage unit 720 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 7201 and / or a cache memory unit 7202 , and may further include a read-only memory unit (ROM) 7203 .

[0268] The storage unit 720 may also include a program / utility 7204 having a set (at least one) of program modules 7205, such program modules 7205 including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0269] Bus 730 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0270] The electronic device 700 may also communicate with one or more external devices 700' (e.g., keyboards, pointing devices, Bluetooth devices, etc.) to enable a user to communicate with the electronic device 700, and / or any device (e.g., routers, modems, etc.) that the electronic device 700 can communicate with one or more other computing devices. Such communication may be performed via an input / output (I / O) interface 750. In addition, the electronic device 700 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 760. The network adapter 760 may communicate with other modules of the electronic device 700 via a bus 730. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 700, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0271] Through the above description of the implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by combining software with necessary hardware. Figure 8 As shown, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes a number of instructions to enable a computing device (which can be a personal computer, a server, or a network device, etc.) to execute the above method according to the embodiment of the present disclosure.

[0272] The software product may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0273] The computer readable storage medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by an instruction execution system, an apparatus, or a device or used in combination with it. The program code contained on the readable storage medium may be transmitted with any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the above.

[0274] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0275] The computer-readable medium carries one or more programs. When the one or more programs are executed by a device, the computer-readable medium implements the following functions: obtaining cargo information and transportation requirements of multiple cargo sources; obtaining driver information and transportation tool information, wherein the driver information includes driver basic information and transportation plan information; determining multiple tags and their weights, wherein the multiple tags include at least a time matching tag and a location matching tag, wherein the time matching tag is used to indicate the matching degree between the driver and the cargo source in the time dimension, and the location matching tag is used to indicate the matching degree between the driver and the cargo source in the location dimension; determining the score of each of the multiple tags based on the driver basic information, the transportation plan information, the transportation tool information, the cargo source information and the transportation demand; calculating the matching score between each cargo source and the driver based on the score of each tag and its weight; and recommending cargo sources to the driver according to the matching score.

[0276] Those skilled in the art will appreciate that the above modules can be distributed in the device according to the description of the embodiment, or can be changed accordingly and only used in one or more devices different from the embodiment. The modules of the above embodiments can be combined into one module, or further divided into multiple sub-modules.

[0277] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the embodiment of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiment of the present disclosure.

[0278] The exemplary embodiments of the present disclosure are specifically shown and described above. It should be understood that the present disclosure is not limited to the detailed structures, configurations or implementations described herein; on the contrary, the present disclosure is intended to cover various modifications and equivalent configurations included in the spirit and scope of the appended claims.

Claims

1. A method for recommending a source of goods to a driver, characterized in that: include: Obtain supply information and transportation requirements of multiple sources; Acquiring driver information and transportation tool information, wherein the driver information includes basic driver information and transportation plan information; Splicing the cargo sources, including: obtaining matching cargo sources that match the "transportation tool information" of the driver, splicing the matching cargo sources that are continuous in time and place to form a spliced ​​cargo source, and adding the spliced ​​cargo source to the cargo source information; Determine a plurality of tags and their weights, wherein the plurality of tags include at least a time matching tag and a location matching tag, wherein the time matching tag is used to indicate the matching degree between the driver and the cargo source in the dimension of time, and the location matching tag is used to indicate the matching degree between the driver and the cargo source in the dimension of location; The score of each of the multiple tags is determined according to the basic information of the driver, the transportation plan information, the transportation tool information, the cargo source information and the transportation demand. Specifically, after obtaining the transportation plan information of the driver, the idle information of the driver is determined according to the transportation plan information. After obtaining the cargo source information and the transportation demand, the loading time and place of the transportation demand of the cargo source and the unloading time and place of the transportation demand are determined. The scores of the time matching tag and the location matching tag are determined according to the idle information of the driver, the loading time and place of the transportation demand of the cargo source, and the unloading time and place of the transportation demand. The time matching tags include two types of time matching tags, wherein the first type of time matching tags is used to indicate whether the time interval of the cargo transportation demand is within a certain idle time interval of the driver, and the second type of time matching tags is used to indicate the length of the time intervals between the two end points of the cargo transportation demand time interval and the two end points of the driver's idle time interval when the cargo transportation demand time interval is within a certain idle time interval of the driver. When the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, the score of the location matching tag is calculated, wherein the location matching tag is used to indicate the distance between the loading location of the transportation demand of the cargo source and the location of the driver at the left end point of the free time interval of the driver, and the distance between the unloading location of the transportation demand of the cargo source and the location of the driver at the right end point of the free time interval of the driver when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver; Calculate the matching score between each cargo source and the driver based on the score and weight of each tag; Recommend cargo sources to drivers based on the matching scores.

2. The method according to claim 1, characterized in that The multiple labels also include at least one of the following: a transport model matching label and a transport load matching label.

3. The method according to any one of claims 1 to 2, characterized in that: The method further includes optimizing the weight of each tag, and optimizing the weight of each tag specifically includes: After recommending cargo sources to the driver according to the matching degree, the driver's evaluation information on the cargo sources is recorded; Optimize the weight of each label based on the driver's evaluation information on the source of goods.

4. The method according to claim 3, characterized in that The step of optimizing the weight of each label according to the driver's evaluation information on the cargo source specifically includes: Adjust the weight of the tag, recalculate the matching score between each cargo source and the driver based on the score of each tag and its adjusted weight, and re-recommend cargo sources based on the matching score; Obtaining drivers' ratings of the re-recommended cargo sources; Calculating a recommended deviation parameter according to a deviation between a matching score between a cargo source and a driver and a score given by the driver to the re-recommended cargo source; The weight of the label when the recommended deviation parameter is the smallest is determined as the weight of the optimized label.

5. A device for recommending cargo sources to drivers, characterized in that: include: A first acquisition unit is used to acquire the supply information and transportation requirements of multiple supply sources; A second acquisition unit is used to acquire driver information and transportation tool information, wherein the driver information includes basic driver information and transportation plan information; a first determination unit, for determining a plurality of tags and their weights, wherein the plurality of tags at least include a time matching tag and a location matching tag, wherein the time matching tag is used to indicate the matching degree between the driver and the source of cargo in the dimension of time, and the location matching tag is used to indicate the matching degree between the driver and the source of cargo in the dimension of location, and the time matching tag includes two types of time matching tags, wherein the first type of time matching tags is used to indicate whether the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver, and the second type of time matching tags is used to indicate the length of time intervals between two end points of the time interval of the transportation demand of the source of cargo and two end points of the free time interval of the driver respectively when the time interval of the transportation demand of the source of cargo is within a certain free time interval of the driver; a second determination unit, configured to determine the score of each of the multiple tags according to the basic information of the driver, the transportation plan information, the transportation tool information, the cargo source information and the transportation demand, specifically, after obtaining the transportation plan information of the driver, determine the driver's free information according to the transportation plan information, after obtaining the cargo source information and the transportation demand, determine the loading time and place of the transportation demand of the cargo source, and the unloading time and place of the transportation demand, determine the time matching tag and the score of the place matching tag according to the driver's free information, the loading time and place of the transportation demand of the cargo source, and the unloading time and place of the transportation demand, and when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver, calculate the score of the place matching tag, wherein the place matching tag is used to indicate the distance between the loading place of the transportation demand of the cargo source and the location of the driver at the left end point of the driver's free time interval, and the distance between the unloading place of the transportation demand of the cargo source and the location of the driver at the right end point of the driver's free time interval when the time interval of the transportation demand of the cargo source is within a certain free time interval of the driver; A calculation unit, used for calculating the matching score between each cargo source and the driver according to the score and weight of each tag; The recommendation unit is used to recommend cargo sources to drivers according to the matching scores. The first acquisition unit further executes the step of splicing cargo sources, including: acquiring matching cargo sources that match the driver's "transportation tool information", splicing matching cargo sources that are continuous in time and place to form a spliced ​​cargo source, and adding the spliced ​​cargo source to the cargo source information.

6. An electronic device, characterized in that: include: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

7. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images